HP Services Software Patches - alpshad12_061

 
NOTE:  An OpenVMS saveset or PCSI installation file is stored
       on the Internet in a self-expanding compressed file.
       The name of the compressed file will be kit_name-dcx_vaxexe
       for OpenVMS VAX or kit_name-dcx_axpexe for OpenVMS Alpha.
 
       Once the file is copied to your system, it can be expanded
       by typing RUN compressed_file.  The resultant file will
       be the OpenVMS saveset or PCSI installation file which
       can be used to install the ECO.
 
Copyright (c) Digital Equipment Corporation 1996.  All rights reserved.

PRODUCT:    Volume Shadowing for OpenVMS Alpha

OP/SYS:     OpenVMS Alpha 

COMPONENT:  Shadow Driver 

SOURCE:     Digital Equipment Corporation

ECO INFORMATION:

     ECO Kit Name:  ALPSHAD12_061
     ECO Kits Superseded by This ECO Kit:  ALPSHAD11_061
                                           ALPSHAD10_061 
     ECO Kit Approximate Size:  1062 Blocks
     Kit Applies To:  OpenVMS Alpha V6.1, V6.1-1H1, V6.1-1H2
     System/Cluster Reboot Necessary:  Yes
     Installation Rating:  2 -  To be installed on all systems running
                                the listed version of OpenVMS and
                                using the following feature:
                                Volume Shadowing.

     NOTE:  In order to receive the full fixes listed in this kit,
            the following remedial kits also need to be installed:

                 ALPSHAD09_061


ECO KIT SUMMARY:

An ECO kit exists for Volume Shadowing on Alpha V6.1 through 
V6.1-1H2.  This kit addresses the following problems: 

Problems Addressed in ALPSHAD12_061:

  o  During a Bad Block Recovery operation, if the source member of
     the shadow set cannot be found by the SHDRIVER, a SHADDETINCON
     system crash will occur.


Problems Addressed in ALPSHAD11_061:

  o  An application I/O doing either a read or a write may fail.
     This can occur even though only one member of a multiple 
     member shadow set is unable to complete the requested I/O.  
     This results in the shadow set members being removed from the 
     set.  The I/O will be returned to the application with the 
     appropriate read or write error.

  o  Due to an incorrect recorded MSCP flag field, segmented I/Os 
     fail with invalid MSCP modifiers errors.

  o  Due to various code problems ,queue corruption, IRPs on 
     inappropriate queues, ACCVIOs, SHADDETINCON bugchecks, 
     and incorrect/inconsistent queue_fl and queue_bl pointers 
     in elements may occur.

  o  Excessive but allowable system process quota limits (PQL
     parameters) can cause a shadow server process to be created
     that is not able to perform the task for which it is intended,
     specifically, shadow set copies and merges.  Depending on 
     the system parameters, the process may exit or it may enter a
     resource wait state and never transition to a useful state.

  o  Many I/Os that complete with errors have extraneous information 
     in the second longword of the IOST/IOSB.  If the I/O is a data 
     transfer I/O, this may be interpreted as a transfer count.  
     Most products investigated see the error and ignore the transfer 
     count.  However, there may be applications that interpret the  
     transfer count as "this many bytes processed before the error 
     occurred" which is incorrect.

  o  A system may crash with a REFCNTNEG bugcheck during shadow set
     copy/merge operations.

  o  A system may crashes with an INVEXCEPTN bugcheck if a shadow set
     member is write-protected and a write to that member is attempted.

  o  A node may hang due to no quorum after losing its connection to
     the quorum disk.

  o  Layered drivers (e.g., Host-Based Raid) error recovery mechanisms   
     require the DUDRIVER interface to allow improved reliability.

  o  If a system disk is mounted in the cluster as a data disk and
     a system is booted from a non-master member, the system will
     crash with a SHADDETINCON crash in SHSB$READ_SCB.

  o  A SHADZEROMBR bugcheck may occur during an attempt to determine  
     if there is a valid source member of the shadow set that could 
     become the master.

  o  If an Alpha system disk is mounted on a different Alpha system  
     as a data disk and the first Alpha crashes and reboots more 
     than once, the Virtual Unit UCB$L_RWAITCNT field on the second
     Alpha system will go negative.  This will hang access to the
     second Alpha's device.

  o  The system will crash with an ACCVIO in SHSB$CANCEL_IRP_CHEC
     if a 'STOP PROCESS/ID=' command is issued to a process that 
     has IO outstanding to a shadow set.

  o  A shadowing crash may occur with the VCB address used as the
     UCB address, when the shadow set has no member with the master 
     bit set in SHAD$B_MBR_STATUS.

  o  A system crash may occur when an attempt is made to mount 
     a write-protected source member.


Problems Addressed in ALPSHAD10_061:

  o  Sometimes a multiple member system disk shadow set does 
     not correctly complete a merge operation.  The merge
     is aborted but the shadow set is marked as though the 
     merge had completed.  If the virtual unit does not exist 
     when the system is booted, this problem will not be 
     encountered.

  o  A SHADZEROMBR bugcheck can occur even though there are
     valid source members in the shadow set.


INSTALLATION NOTES:

In order for the corrections in this kit to take effect, the system
must be rebooted.  If the system is a member of a VMScluster, the 
entire cluster should be rebooted. 

Files on this server are as follows:

»alpshad12_061.README
»alpshad12_061.CHKSUM
»alpshad12_061.CVRLET_TXT
»alpshad12_061.a-dcx_axpexe